Stroke and cognitive impairment remain a major public health problem with a disproportionate impact on blacks and Hispanics. The reasons for these race-ethnic disparities are not entirely clear. Improved detection and control of novel stroke risk factors are needed to reduce vascular disease burden. The Northern Manhattan prospective cohort consists of a population-based cohort of 3298 stroke-free adults enrolled since 2001 who have been followed annually for stroke, MI, and death. This collaborative study now involves faculty across 2 academic institutions and is the first prospective cohort study among whites, blacks and Caribbean Hispanics living in the same community. Over the last 5 years, the aims have expanded to include subclinical carotid, cardiac, and brain measures. High-resolution carotid imaging data has been assembled on 1770 subjects and a neuropsychological battery and standardized brain MRI will be completed on 1300 subjects with quantitative analyses to measure white matter hyperintensities, silent infarcts, and silent cerebral microbleeds. This grant supports continued follow-up and outcome detection and expansion of data collection in this cohort. The aims are to evaluate the relationship between vascular outcomes (stroke, MI, and vascular death) and adiponectin, a new marker of obesity, new measures of subclinical carotid disease (carotid plaque area and densitometry), and quantitative MRI measures of vascular subclinical brain disease, as well as evaluate the impact on cognitive decline. To accomplish these aims, stored baseline plasma will be used to measure fasting adiponectin levels. High-resolution carotid imaging data will be quantitatively analyzed from standardized digital recordings. Repeat neuro-psychological testing emphasizing frontal-executive domains will be done on 1105 subjects. Subjects will be followed by annual telephone interviews to ascertain stroke, MI, death, and changes in cognitive state. In-person assessment will be done for all subjects who screen positive for any vascular event. Community stroke surveillance will be maintained to insure stroke detection among the cohort. The strengths of this cohort study are the wealth of baseline data already assembled, the tri-ethnic composition residing in the same community, the outstanding follow-up record, the evaluation of traditional and emerging risk factors, and the innovative assessment of MRI subclinical disease and carotid subclinical disease.